Accurate remaining useful life prediction of turbofan engines can effectively avoid serious air-disaster due to engine failure by mining its components degradation characteristics. However, the complexity of engine components degradation characteristics keeps increasing when the airplane flies in complex environments and multi-operating points (MOP) mode. In this work, a deep learning fusion algorithm based on self-attention mechanism (SAM) is proposed. This algorithm uses one-dimensional convolutional neural network (CNN) to extract the spatial features and long-short-term memory networks (LSTM) to fuse the measurement data of 21 components and extract the temporal feature from the measured data. Furthermore, with extracted features and SAM, the proposed algorithm provides weight redistribution and solves the information loss problem in LSTM. Experimental results validated the proposed model and it is found that the proposed prediction model can predict RUL of turbofan engines accurately and stably under MOP mode and the proposed model outperforms other latest existing approaches.

Photovoltaic (PV) modules are exposed to the outside, which is affected by radiation, the temperature of the PV module back-surface, relative humidity, atmospheric pressure and other factors, which makes it difficult to test and analyze the performance of photovoltaic modules. Traditionally, the equivalent circuit method is used to analyze the performance of PV modules, but there are large errors. In this paper—based on machine learning methods and large amounts of photovoltaic test data—convolutional neural network (CNN) and multilayer perceptron (MLP) neural network models are established to predict the I–V curve of photovoltaic modules. Furthermore, the accuracy and the fitting degree of these methods for current–voltage (I–V) curve prediction are compared in detail. The results show that the prediction accuracy of the CNN and MLP neural network model is significantly better than that of the traditional equivalent circuit models. Compared with MLP models, the CNN model has better accuracy and fitting degree. In addition, the error distribution concentration of CNN has better robustness and the pre-test curve is smoother and has better nonlinear segment fitting effects. Thus, the CNN is superior to MLP model and the traditional equivalent circuit model in complex climate conditions. CNN is a high-confidence method to predict the performance of PV modules.

Short-term traffic flow prediction (STFP) is one of the key technologies in Intelligence Transportation System (ITS). With the development of artificial intelligence technology, deep learning has been employed to STFP and certain achievements have been obtained. However, further improvement of STFP accuracy is constrained by the presence of noise, modal aliasing in the modal components and insufficient feature extraction of neural networks. Therefore, to more accurately capture the changing trend of traffic flow in different time periods, and increase the reliability and interpretability of the prediction models, improved variational mode decomposition (IVMD) and self-attention mechanism (SAM) based hybrid convolutional neural network (CNN) and long short term memory network (LSTM) are proposed for STFP in this study, benefited from the modal decomposition of IVMD and redistribution of the neural weights in CNN-LSTM networks by SAM, the proposed IVMD-CNN-LSTM-SAM effectively suppresses the influence of randomness and volatility of traffic flow, and accurately predict the short-term lane occupancy in the designated area. Furthermore, this study analyses the mechanism of SAM and IVMD in STFP, and proves the superiority of IVMD and SAM in signal decomposition with physical significance and weight allocation respectively. Therefore, the interpretability of the proposed neural network model is significantly improved. Eleven competitive neural networks forecasting methods are used as benchmarks and numerical studies based on actual traffic flow validate the accuracy and stability of the proposed IVMD-CNN-LSTM-SAM. Mean absolute percentage error and goodness of fit of the proposed model are 0.81% and 0.9968, respectively.

Accurate photovoltaic (PV) power prediction is of great significance for the stable operation of PV system, but the PV power sequence is nonstationary, so it is difficult to establish the prediction model effectively by a simple neural network. In this study, the MCVMD-MI-SWATS-Codec (multidimensional constraints variational mode decomposition-mixed initialization-switching from Adam to stochastic gradient descent-codec) that is based on the idea of deep model fusion is proposed to predict PV power generation. MCVMD method with parameter K determined by multidimensional constraint criterion is used to decompose the PV power data, and the frequency of each component sequence is analyzed after decomposition to explore the physical characteristics and application value of each component frequency. Then, a hybrid ResNet-LSTM (residual network-long- and short-term memory) model based on codec mechanism integrates input data with different dimensions, such as weather conditions and historical IMF (intrinsic mode function), into dense vectors with the same dimension. The experimental data of polysilicon PV array in the Australian desert environment are used to test the proposed fusion neural network model and the other six competitive models. The results show that MCVMD algorithm is significantly helpful in decomposing the nonstationary data to improve the prediction accuracy, and MCVMD-MI-SWATS-Codec model has high prediction accuracy and robustness in both stable and unstable weather conditions.